Institutional Repository of Vilnius University
Maisto leksika dvikalbiame lietuvių–anglų kalbų žodyne ir ChatGPT sugeneruotuose žodyno straipsniuose /
Abstract
dc:descriptionThis bachelor‘s thesis aims to describe the presentation of non-equivalent Lithuanian food and drink lexis in a bilingual Lithuanian–English Dictionary (2006; Lith. Didysis lietuvių–anglų klabų žodynas, DLAKŽ) and in dictionary entries generated by ChatGPT. The aim of the thesis is to evaluate the quality of ChatGPT generated dictionary entries and compare them with the entries provided in DLAKŽ. The research addresses the question of whether AI generated dictionary entries could be transferred to a bilingual dictionary or used to supplement one. The theoretical part of the thesis discusses the concept of non-equivalent lexis, its presentation in bilingual dictionaries, the concept of DLAKŽ. The empirical part analyses the dictionary entries of 30 Lithuanian food and drink names. Using qualitative linguistic analysis and the comparative method, the study evaluates descriptive glosses, grammatical information, stylistic and usage labels, and illiustrative examples. The study shows that ChatGPT often provides more detailed descriptive glosses and includes illustrative examples that reveal cultural context, however, their quality raises concerns. Innacurate labels, redundant information, stress-marking errors, and occasional hallucinations occur. Therefore, it can be argued that ChatGPT generated entries may supplement a lexicorgapher‘s work but cannot replace professionaly prepared dictionary entries. Thesis is relevant to lexicographers, translators, artificial intelligence specialists , and the general public.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Bieliūnaitė, Ieva,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD308317703&prefLang=en_US